Using data from the national micronutrients survey 2011–2012, the present study explored the status of subclinical vitamin A nutrition and the underlying determinants in the Bangladeshi population.

A nationwide cross-sectional study.

The survey covered 150 clusters; fifty in each of rural, urban and slum strata.

Three population groups: (i) pre-school age children (6–59 months; PSAC); (ii) school age children (6–14 years; SAC); and (iii) non-pregnant non-lactating women (15–49 years; NPNLW).

National prevalence of subclinical vitamin A deficiency was 20·5, 20·8 and 5·3 % in PSAC, SAC and NPNLW, respectively. Slum populations had higher prevalence compared with urban (PSAC: 38·1 v. 21·2 %, P<0·001; SAC: 27·1 v. 22·1 %, P=0·004; NPNLW: 6·8 v. 4·7 %, P=0·01). Dietary vitamin A met up to 27·1–46·0 % of daily needs; plant-source vitamin A constituted 73–87 % of the intakes. Multivariable regression analyses showed that higher consumption of animal foods was associated with higher retinol status in PSAC (β=0·27; P<0·001); and living in urban area was related to higher retinol status in NPNLW (β=0·08, P=0·004) and PSAC (β=0·11, P=0·04). Increased intake of leafy vegetables was associated with lower retinol status in SAC (β=−0·08, P=0·02). Vitamin A supplementation in PSAC did not significantly influence serum retinol within one year post-supplementation (P>0·05 for differences in β between <3 months v. 3–6 months, 6–9 months and 9–12 months).

Prevalence of subclinical vitamin A deficiency was high in children in Bangladesh. Intakes of animal-source foods and leafy vegetables were associated with higher and lower retinol status, respectively. Increased food diversity through animal-source foods is required.